Parkinson's disease (PD) is a common and disabling condition in the expanding elderly population of the US and worldwide. Its etiology remains unknown and both genetic and environmental factors have been suspected. The long-term goal of the proposed studies based in different sampling is to clarify the etiology of PD and to identify means to prevent it. Three independent but related studies based on different sampling and measurement strategies are proposed. The hypotheses tested derive directly from our current work and preliminary findings. A case control study will include 800 cases of PD referred to the Mayo clinic from a 120 mile radius or from a 5 state region and 800 controls free pf PD and parkinsonism matched by age (+ 2 years), sex and region of residence. Controls from the general population will be identified from health care financial administration lists for cases aged 65 years or above and through random digit dialing for cases below 65 years. Exposures will be accessed through direct telephone interview, and will include tobacco, coffee, and alcohol use; markers of novelty seeking behavior; and, for women, use of estrogen replacement therapy after menopause and other reproductive and estrogen related factors. A first historical cohort study will test the association between unilateral and bilateral oophorectomy before menopause and PD in an established population based cohort. The study will include 2,533 women who underwent oophorectomy in 1950-1987 while residing in Olmsted County, MN and 2,533 women of the same age and residence who did not undergo oophorectomy. A second historical study will test the association between personality traits measured by the Minnesota Multiphasic Personality Inventory (MMPI) and PD in an established research cohort. This study will include 8,775 persons who underwent MMPI testing in 1962-1965 while residing in Minnesota. The proposed case-control study is strong because it has adequate statistical power to confirm our preliminary findings on the role of estrogen in PD and to explore the link between substance use and novelty seeking behaviors in PD. All interviews with case and controls will be direct, the proposed oophorectomy cohort study is strong because of its cohort design, its population-based sampling, its adequate statistical power, and because of the expected high rate of follow-up through both passive and active strategies. The proposed MMPI cohort study is strong because of its cohort design, its adequate statistical power and because of our extensive experience with tracing and interviewing individuals. These three studies will contribute greatly to understanding the causes and possible prevention of PD by exploring novel hypotheses and by using innovative methods.